Ant Control

Ant activity is a moisture signal before it is a pest problem — colonies exploit damaged wood, failed flashing and wet sill plates.

Why are carpenter ants in a house a moisture problem before they are a pest problem?

Because carpenter ants do not eat wood. They chew it out and throw it away.

That single fact reorganises everything about how the problem should be handled. A termite consumes cellulose and can therefore work happily in sound, dry structural timber. A carpenter ant has no such capacity — it is an omnivore that forages for sugars and protein, and the wood it removes is not food but real estate. It is excavating a nest cavity. And because excavating sound timber is expensive work for an insect with a set of mandibles and nothing else, colonies overwhelmingly choose timber that something has already softened for them.

That something is almost always water.

So when a homeowner in Northport or Cold Spring Harbor finds a line of large black ants crossing a bathroom floor in April, the useful question is not “what do I spray.” It is “what is wet.” Somewhere in that structure there is a member holding moisture — a sill plate sitting on a foundation with no capillary break, a band joist behind a deck ledger that was flashed badly or not at all, a window sill under a leaking exterior casing, the framing around a shower pan that failed a decade ago, the ceiling joist under a chimney flashing that lets water in during driving rain from the north.

The ants found it before you did. In practical terms they are a moisture meter with legs.

This is why our first hour on an ant call generally has very little to do with insects. We are looking at gutters and where they discharge, at the grade around the foundation, at the flashing at roof-to-wall junctions, at crawlspace ventilation and vapor barriers, at the condensate line from the air handler, at every plumbing wall. When we find the ants, we usually find them at a place we had already flagged for a different reason.

And it is why an ant proposal from us sometimes includes items that read like carpentry. Rebedding a length of flashing, extending a downspout, replacing a rotted sill section — that is ant work. It removes the condition rather than the current occupants.

How do you tell a carpenter ant from a termite?

By the waist, the antennae, the wings and the debris. Any one of these will usually settle it; together they are conclusive.

The waist. An ant of any species has a sharply constricted, pinched waist between thorax and abdomen — a distinct narrowing you can see with the naked eye on an insect as large as a carpenter ant. A termite has no waist at all. Its body is broad and continuous from front to back, more like a grain of rice than an hourglass.

The antennae. Ant antennae are elbowed: a long first segment, then a sharp bend, then the rest. Termite antennae are straight and beaded, like a tiny string of pearls.

The wings. This matters most in spring, when both insects produce winged reproductives and both get noticed at the same windows. A winged carpenter ant has two pairs of wings of clearly unequal length — the front pair distinctly longer than the hind pair. A termite swarmer has four wings of equal length, roughly twice the length of its body, and they detach readily. A scatter of identical discarded wings on a windowsill, in a spider web, or along a basement slab is a termite signal, not an ant one.

The debris. This is the field diagnostic that decides most cases. Carpenter ants push shavings out of their galleries, and those shavings are coarse and fibrous — recognizably wood, often with fragments of dead insects and other colony refuse mixed in. Termites do not produce anything of the kind. Subterranean termites, the species established across Long Island, build shelter tubes of soil and fecal material to bridge from the ground to the timber, and their galleries are packed with that same mud.

So: clean, smooth-walled galleries with sawdust below them means ants. Soil-packed galleries with mud tubes on the foundation means termites.

The distinction is worth getting right because the two answers do not lead to the same place, or even to the same firm. Ant work is colony work, and it is what this page is about. Termite work is soil and structure work, and Graduate does not do it — which is precisely why the identification detail sits here in full rather than being waved at. Confusing the two costs a season, and with termites a season is expensive. The longer comparison, with the shed wings, the tubes and the way the damaged wood reads when you open it, is set out in the eastern subterranean termite profile.

What is frass, and how do you read it?

Frass is the excavation spoil — the material a carpenter ant colony pushes out of the gallery it is cutting — and it is the most useful piece of evidence on an ant job.

It looks like pencil sharpenings rather than dust. Individual particles are shredded fibers of wood, pale, irregular, and often mixed with the discarded body parts of insects the colony has fed on and the shed skins of its own brood. That biological content is what distinguishes carpenter ant frass from ordinary sawdust left by a contractor, and it is also what distinguishes it from the fine, flour-like powder that wood-boring beetles produce.

The critical thing about frass is where it lands. Ants push it out through a small gallery opening — often a slit no wider than a pencil lead, frequently in a seam that reads as a normal construction joint. The spoil then falls straight down. So a small drift of frass on a windowsill, on the top of a baseboard, on a basement floor beneath a joist, in the corner of a garage, or on the flat of a porch beam is a vertical pointer to a gallery immediately above it.

This is much better information than watching foragers. A foraging carpenter ant will travel a considerable distance from its nest, along a fence, up a downspout, across a deck rail, in through a gap at a window casing. Following ants tells you their route; finding frass tells you their address.

We also read the frass itself. Fresh spoil is pale and loose. Older accumulations gray off, compact and pick up dust. A pile that reappears within days of being swept is an active gallery being actively enlarged. A single gray deposit that does not return may be historical — a colony that has already moved or died — and we would rather establish that than treat a wall for nothing.

Where there is frass but no obvious opening, the gallery is usually behind the finish surface, and the sound of it can help. A colony working inside a hollow void produces a faint dry rustling that is audible against a quiet wall at night, and tapping the member will often change the pitch where the wood behind it has been hollowed out.

What is the difference between a parent colony and a satellite colony?

A parent colony holds the queen and the eggs. Satellite colonies hold everything else — older larvae, pupae, winged reproductives and workers — and they can be a considerable distance away. Understanding that split is the difference between resolving a carpenter ant problem and chasing it around a house for two years.

The parent nest needs consistently high moisture, because eggs and young larvae will not survive drying out. That is why parent colonies are usually outdoors: in a dead tree, a stump, a buried root, a stack of firewood, a landscape tie, a fence post, a wooden retaining wall. On a wooded North Shore property there may be several within a hundred feet of the house and none of them visible from the driveway.

Satellite nests are far less demanding. Mature brood tolerates drier conditions, so a satellite can establish in a wall void, a ceiling cavity, the space above a bay window, the hollow of a porch column, behind foam insulation, inside a hollow door, in the void behind a dishwasher or under a shower base. Some of these sites contain very little wood at all — the ants are using an existing cavity rather than cutting one.

Traffic runs constantly between the two. That means a house can be genuinely infested — ants indoors, frass, activity in a wall — while the queen sits in a stump beyond the property line. Kill the satellite and the parent colony repopulates it. Kill the foragers and the parent colony sends more.

It also means the treatment has to travel. This is why bait, correctly placed and correctly matched to what the colony currently wants to eat, does work that a contact spray cannot: it is carried home. Workers share liquid food with one another and with brood by trophallaxis, and a slow-acting material moves through the colony network — including out to a parent nest we have never seen — before it takes effect. A fast knockdown does the opposite. It kills the courier before the parcel is delivered.

Where we can locate a satellite precisely, direct treatment into that void is appropriate and quick. But locating it precisely is the operative phrase, and it is a job for frass, sound and thermal reading rather than for guesswork.

Which other ants actually show up in New York City and Long Island buildings?

Carpenter ants get the headline because they interact with the structure. They are not the most frequent call. Two other species account for most indoor ant complaints in this region, and neither responds well to the treatment that suits the other.

Odorous house ants. Small, dark brown to black, and named for the smell they release when crushed — commonly described as rotten coconut or blue cheese. They are the most difficult of the common indoor ants, for a structural reason rather than a chemical one. Their colonies have many queens and no fixed single nest. Instead they occupy a network of small nest sites — under a slab, in a wall void, in mulch, in a potted plant, under a doormat, in insulation — and they relocate readily. Crucially, they respond to a disturbance by budding: workers and queens split off and establish new sites elsewhere. So an aggressive contact spray on an odorous house ant trail is not just ineffective, it actively multiplies the problem, turning one network into three.

The correct approach is patient. Do not disrupt the trail. Bait it, let the trail carry the material, and accept that the population may rise briefly before it collapses because you have recruited more workers to the food. Then find what is drawing them: they are strongly attracted to moisture and to honeydew from aphids and scale on shrubs planted against the house, which is why cutting foundation plantings back off the siding does more for a recurring odorous house ant problem than another quart of anything.

Pavement ants. Small, brown-black, often with faint parallel grooves on the head and thorax. They nest under slabs, walkways, patios, driveway aprons and against foundations, and they push out characteristic small craters of excavated soil along the joints in a sidewalk or a garage floor. Indoors they typically appear where a slab meets a wall, at an expansion joint, along a basement perimeter or around a sunken tub. They are less mobile than odorous house ants and less structurally destructive than carpenter ants, and the work is mostly about identifying where the trail crosses from soil to interior and closing or treating that transition. In a city building with a slab-on-grade cellar and old expansion joints, that transition can be the entire perimeter.

There are others we see — pharaoh ants in institutional and multi-unit buildings, which absolutely must not be sprayed because they bud violently; acrobat ants, which frequently move into voids that carpenter ants or decay have already opened, and which therefore carry the same moisture message; field and cornfield ants that swarm on a patio in late summer and cause alarm without ever entering the building. Identification decides the method. That is not a formality.

Why does spraying the perimeter stop working?

Because a perimeter band treats the wrong population, and because for some species it makes matters worse.

Consider what a residual barrier around the foundation actually does. It intercepts insects crossing a treated surface. The insects crossing it are foragers — the fraction of the colony out looking for food, which for most ant species is a modest share of the total headcount and one the colony replaces continuously. The queen never crosses it. The brood never crosses it. So the barrier reduces the number of ants you personally see while leaving the reproductive core untouched, and the moment the residue weathers off — sun, rain, irrigation and the ordinary lifespan of any exterior deposit — the count returns.

That is the mechanical failure. There are three more.

It does not reach an indoor nest. If the satellite colony is in a wall cavity above the bathroom, treating the outside of the house is treating a place the colony does not need to go.

It does not address the attractant. The wet band joist is still wet. The scale insects on the yews against the wall are still producing honeydew. The colony’s reason for being there has not changed, and something will keep arriving.

With budding species it fragments the colony. As above: pressure a multi-queen colony with a repellent material and it splits. You have converted one problem into several, distributed further into the building, and the visible result is often worse three weeks later.

None of this means exterior work is useless. A well-targeted exterior application at the right time of year, combined with bait and with correction of the conditions, is a legitimate part of the job. The failure is treating the perimeter as the whole of the job. Ryan puts it plainly: if you want someone to spray and leave, we are not the right fit.

How do you find the moisture path that is feeding the colony?

By following water backwards from the wet timber, and by accepting that the entry point for the water is frequently nowhere near the damage.

Start at the gallery. The frass locates the gallery; the gallery locates the wet member. That member is the downstream end of the path, not the source.

Look up and outboard. Water moves down and sideways through a structure along framing, sheathing and behind cladding, often traveling several feet horizontally before it appears. A wet band joist on the east wall can be fed by a gutter that overshoots at one corner. A soaked window sill can be fed by a siding butt joint six feet above it. A rotted porch beam is usually fed by the deck surface above it, not by rain landing on the beam.

The usual suspects, in the order we find them. Gutters that are undersized, blocked, pitched wrong, or discharging within a foot of the foundation. Grade that slopes back toward the house, which is extremely common on properties where landscaping has been rebuilt over the decades and the bed level has crept up above the sill. Missing or reverse-lapped flashing at roof-to-wall junctions, at deck ledgers, above windows and doors, and at chimney saddles. Deck ledgers bolted straight to a band joist with no flashing at all — a common detail on 1970s and 1980s construction and a reliable carpenter ant site. Crawlspaces with no vapor barrier over bare soil, which drive the humidity of the entire floor system above them. Condensate lines that discharge into a wall or under a slab. And slow supply-line leaks behind a dishwasher, a refrigerator ice line or a shower valve, which produce a small, permanently damp area that a satellite colony finds ideal.

Confirm, do not assume. Moisture meters read what a member actually holds rather than what it looks like it holds, and painted trim can be sound-looking and saturated. Probing a suspect member with a tool tells you where the soft zone starts and stops. A borescope through a small hole shows whether a void is occupied before anything is opened up.

The point of all this is to write a scope that fixes the cause. The colony is a symptom, and it is generally the least expensive part of the situation to resolve. The wet framing is the actual liability, and it will produce decay whether or not any insect ever finds it. This is the same reasoning that drives our structural exclusion work, where the building’s condition rather than its current occupants is the object of the exercise.

When in the year do ants appear, and what does the timing tell you?

The calendar is diagnostic. Where ants show up, and when, narrows the search before anyone opens a wall.

Late winter and early spring. Carpenter ants indoors between January and March are the most informative sighting of the whole year. A colony living outdoors in a stump is inactive in cold weather — it cannot forage across frozen ground and it is not walking through your kitchen. So ants moving indoors in midwinter are living inside the heated envelope. There is a satellite nest in the structure, and the warmth of the house has convinced it that spring has arrived. Nobody wants to hear this in February, but it is far better information than a sighting in June.

Spring. General emergence. Overwintered colonies resume foraging as soil temperatures rise, and this is also when winged reproductives appear. Large winged carpenter ants inside a house — particularly gathering at a window, which is where flying insects head for light — indicate a mature colony that has been established for several years, because colonies do not produce reproductives until they reach size. Spring is also when a termite swarm would occur, hence the identification section above; the two events overlap and are routinely confused.

Early summer. Colony growth accelerates and foraging range expands. Odorous house ants and pavement ants become the dominant call. Heavy rain drives ants up out of saturated soil and into slabs and foundations, which is why an ant complaint frequently follows a wet week rather than a hot one.

Late summer. Colonies shift dietary preference as brood production changes, moving between sweet and protein-rich foods. This is a real operational detail rather than trivia — a bait that was being taken enthusiastically in June can be walked past in August, and the answer is to change the bait rather than to conclude that bait does not work.

Autumn. Foraging pushes indoors as outdoor food declines and temperatures drop. Colonies that will overwinter in a structure position themselves now, which makes autumn a strong time to close gaps at the sill, at utility penetrations, and where foundation plantings touch the siding.

Across all of it, the moisture story runs underneath. A property with a chronic drainage problem produces ants on a schedule; correcting the drainage changes the schedule permanently.

What does our ant work actually include?

Four things, in a fixed order, and the first one takes the longest.

Identification. Species first, because the method depends on it entirely and because a treatment appropriate to a carpenter ant colony can make an odorous house ant infestation worse. We collect specimens rather than working from a phone description, and we take note of size variation within the sample — carpenter ant workers are polymorphic, with a wide range of sizes in a single colony, which is itself a useful identification cue.

Inspection and source location. The full evidence survey: trails and their direction of travel, frass and what is above it, moisture readings on suspect members, the exterior water management, vegetation contacting the structure, and the interior areas where activity has been reported. Where evidence points into a void, we establish what is in the void before we treat it.

Treatment matched to the biology. For carpenter ants, that is generally a combination of targeted application into located galleries and voids, non-repellent exterior work where it is appropriate, and bait placed on active trails and refreshed as the colony’s preference shifts. For budding species, bait-led work and a deliberate avoidance of anything repellent. For pavement ants, work concentrated at the soil-to-structure transition. Materials are chosen for the site and the occupants, and we say plainly what is being used and where.

Correction and exclusion. The moisture repair, the drainage change, the vegetation cut back off the wall, and the sealing of the trail entry points. This is the part that determines whether we are back next year. Where the sealing work is extensive it is scoped as exclusion in its own right; where it is a handful of penetrations it belongs in the ant job.

Follow-up is part of the sequence rather than an upsell. Bait-led work needs to be checked — placements consumed, trails abandoned, frass no longer reappearing — and a colony that has budded needs to be found rather than assumed.

When is ant work not the right answer, and what are the limits?

There are situations where treating ants is the wrong first move, and it is worth saying so.

When the real finding is rot. Sometimes the inspection turns up a sill plate or a band joist that is failing structurally. At that point the ants are a minor line item and the correct next call is to a carpenter or a structural engineer. We will treat what is there and tell you what we found, but we are not going to pretend an insect problem is the reason the porch is sagging.

When the parent colony is off the property. If the parent nest is in a neighboring yard, a street tree or a wooded strip we have no access to, the honest position is that the property will face ongoing pressure and the work becomes about making the structure unattractive and impenetrable rather than about eliminating a colony we cannot reach.

When the species is being misread. Winged insects in spring get identified as ants by homeowners more often than not, and some of them are termites. Getting that wrong in either direction is costly. We would rather examine a specimen than answer the question over the phone.

When it is not an ant. Small dark insects at a window in spring are frequently something else entirely, and the treatment for something else entirely is not ant treatment.

We also do not claim that a single visit ends a carpenter ant problem. It sometimes does. More often the colony network takes a few weeks to collapse through bait, and the moisture correction runs on the repair schedule of whoever is doing the carpentry. Telling a client that up front is more useful than telling them what they would prefer to hear.

Which ant problem has its own page?

One of them, and it is the one that interacts with the structure. Carpenter ants are the wood-nesting species in this region, and the work they need diverges far enough from the rest of the ant calendar to warrant a page of its own: how the colony is split across a property, the night survey that locates the part of it holding the queen, and the water repair that decides whether any of it recurs. That is set out under carpenter ant control.

The other species described above are handled inside this service rather than on separate pages, because the method for each is a variation on baiting and source location rather than a different discipline.

Where does ant work sit in the rest of the practice?

Ant work is a moisture investigation with an insecticide component, which puts it close to several other things we do. The full range of services is set out on the pest control services page.

The nearest neighbor is structural exclusion, since the gaps ants trail through at sills, utility penetrations and siding terminations are the same gaps that admit everything larger, and the sealing half of an ant job is that discipline wearing a different name. On properties where the drainage is the underlying fault, the standing water that keeps the framing wet is frequently the same water driving mosquito pressure in the yard, so the two surveys tend to find each other. And where the wet framing has already opened a gap at the sill line, the rodent control findings usually arrive in the same report.

The adjacent problem we do not take is termites. They follow the same moisture and soil-contact logic into the same members, and spring is the week they get confused with this insect at the same windowsill, but termite treatment is not work this firm does. The eastern subterranean termite profile is there to settle which one you have before anyone quotes anything.

Our written material on the water behind this insect, from leader discharge to crawlspace humidity to a deck ledger nobody flashed, is gathered in the moisture and wood cluster on our blog.

Graduate Pest Control has been working on buildings across Nassau, Suffolk, Manhattan, Brooklyn and Queens since 1983, and the founding view has not changed: we treat every job as a building problem, not a pest problem. If you have ants and sawdust and you would like to know what is actually wet, get in touch — or read more about how Ryan Katz approaches this work.